CN114113516A - Water quality abnormal data detection method based on GAN - Google Patents
Water quality abnormal data detection method based on GAN Download PDFInfo
- Publication number
- CN114113516A CN114113516A CN202111258369.9A CN202111258369A CN114113516A CN 114113516 A CN114113516 A CN 114113516A CN 202111258369 A CN202111258369 A CN 202111258369A CN 114113516 A CN114113516 A CN 114113516A
- Authority
- CN
- China
- Prior art keywords
- water quality
- data
- quality data
- generator
- discriminator
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/18—Water
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Abstract
The invention provides a GAN-based water quality abnormal data detection method, which comprises a water quality data preprocessing method, namely a Principal Component Analysis (PCA) method, wherein the PCA method is used for processing original water quality data, reducing data dimensionality on the premise of not losing original information to the maximum extent, reducing calculated amount and accelerating monitoring time; training the GAN to enable the false normal data generated by the generator to reach a level of falseness, and enabling the discriminator to reach a level of completely discriminating true and false data; and anomaly detection, namely using the trained generator and the trained discriminator for anomaly detection, comparing the data generated by the generator with the original data to obtain reconstruction loss, discriminating the original data by the discriminator to obtain discrimination loss, adding the two losses to obtain a total loss value, and judging whether the original water quality data is normal or not according to the total loss value.
Description
Technical Field
The invention belongs to the technical field of big data analysis and application and artificial intelligence, relates to water quality analysis by using big data, and particularly relates to a water quality abnormal data monitoring method based on GAN.
Background
The water quality detection work is an important part of environmental management, along with the application of the sensor, the water quality detection work is compared with the traditional manual detection, the time efficiency of the water quality detection work is greatly improved, but the problems of low speed, low accuracy and the like still exist at the present stage by how to timely and effectively judge whether the water quality is normal or not through data obtained by the sensor transmission. The generated time series data is detected by a sensor to continuously detect the water quality environment to detect the abnormality. However, most of the conventional anomaly detection methods are based on a probability statistics method, a distance-based method, and a linear model method, and these methods cannot process time-series data well and cannot consider the correlation in a time step, so that they are not suitable for anomaly detection of water quality data.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention aims to provide a GAN-based water quality abnormal data detection method, which captures the dependency on the time step by using an LSTM-RNN, thereby considering the time sequence correlation on the time sequence data detected by a sensor and improving the efficiency of water quality abnormal data detection.
In order to achieve the purpose, the invention adopts the technical scheme that:
a water quality abnormal data detection method based on GAN comprises the following steps:
step 1, preprocessing original water quality data and normal water quality data, screening and processing irrelevant interference data, and reducing data dimensionality;
step 2, training GAN, wherein the GAN consists of a generator and a discriminator, the generator and the discriminator are in a mutual game state, the generator and the discriminator are trained by utilizing the preprocessed original water quality data and the normal water quality data, the generator generates pseudo normal water quality data according to the characteristics of the preprocessed original water quality data, and the discriminator judges the pseudo normal water quality data according to the preprocessed normal water quality data;
and 3, performing anomaly detection on the water quality data to be detected by using the GAN obtained by training in the step 2, comparing the pseudo-normal water quality data generated by the generator with the water quality data to be detected to obtain reconstruction loss, identifying the water quality data to be detected by using the identifier to obtain identification loss, performing accumulation operation on the two losses, and judging whether the water quality data to be detected is anomalous data.
In the step 1, the pretreatment method comprises the following steps:
firstly, carrying out primary pretreatment of data replacement and filling on original water quality data or normal water quality data through Python software;
secondly, removing data with missing key fields in the original water quality data or the normal water quality data;
and finally, carrying out PCA operation to reduce the data dimension.
In step 2, the method for training GAN is as follows:
sending the preprocessed original water quality data into a generator, sending the preprocessed normal water quality data into a discriminator, learning and generating by the generator to obtain pseudo normal water quality data only containing partial water quality data characteristics, learning by the discriminator to obtain the characteristics of the normal water quality data, inputting the pseudo normal water quality data into the discriminator to perform discrimination and judgment operation, and repeating iteration to finish training to obtain the trained generator and discriminator.
In step 3, the abnormality detection method is as follows:
firstly, carrying out PCA operation on water quality data to be detected to reduce data dimensionality;
secondly, respectively inputting the water quality data to be detected after the data dimension is reduced into a generator and a discriminator, mapping the data in the generator to obtain a hidden space, generating pseudo-normal water quality data by the generator according to random variables in the hidden space, and comparing the pseudo-normal water quality data with the water quality data to be detected to obtain reconstruction loss; in the discriminator, directly carrying out discrimination judgment operation to obtain discrimination loss;
and finally, performing accumulation operation on the reconstruction loss and the identification loss to obtain a final loss value, judging whether the water quality data is abnormal according to the loss value, and if the loss value is smaller than a preset threshold value, determining the water quality data as normal water quality data, otherwise, determining the water quality data as abnormal water quality data.
Compared with the prior art, the invention has the beneficial effects that: aiming at the problems of low accuracy, incapability of considering correlation on data time relation and the like in the conventional anomaly detection method, the invention adopts a preprocessing method facing mass water quality detection data to reduce data dimension and improve detection speed; an LSTM-RNN cyclic neural network is adopted to capture the correlation of the water quality data on the time step length and is embedded into a GAN frame, so that the abnormal detection accuracy of the water quality data is improved; abnormal data detection is performed by using GAN. By adopting the three methods, the problems of low accuracy, low speed and the like in the existing water quality data detection method are well solved, and the abnormity detection efficiency of the water quality data is improved.
Drawings
FIG. 1 is a system flow diagram according to an embodiment of the present invention.
FIG. 2 is a diagram of the accuracy of water quality abnormality detection in the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be further described in detail by specific embodiments with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Fig. 1 shows a schematic flow chart according to an embodiment of the present invention, which includes the following steps:
step 1, preprocessing the existing original water quality data and the normal water quality data to screen and process irrelevant interference data and reduce data dimensionality.
Specifically, a pretreatment method for mass water quality data is used, and the steps are as follows:
step 1.1, because the subsequent analysis and detection are all established on the key characteristics of the water quality data, the original water quality data or the normal water quality data are subjected to preliminary preprocessing operations such as data replacement, filling and the like through Python software, for example, NULL data in the original water quality data is replaced by the average value of the data in the row.
Step 1.2, because the subsequent abnormal detection is established on a core field of the water quality data, removing data with missing key fields in the original water quality data or the normal water quality data, for example, detecting missing of some water quality data is more, namely, considering that the data of the corresponding key fields are missing, and deleting the data;
and step 1.3, the water quality data indexes detected by the sensor are various, such as a Ph value, turbidity, dissolved oxygen, water temperature, conductivity, permanganate index and the like. Because the quality of water data sample is constantly producing, the data bulk is huge, and the detection index is more simultaneously, but has certain implication between each item index to quality of water influence degree, through PCA operation, can reduce the data dimension, the efficiency of anomaly detection accelerates, provides the data basis for subsequent anomaly detection.
And 2, training the GAN by utilizing the preprocessed data, so that whether the water quality data is normal or not can be judged. The GAN is composed of a generator and a discriminator, wherein the generator and the discriminator are in a mutual game state, the generator and the discriminator are trained by utilizing the preprocessed original water quality data and the normal water quality data, the generator generates pseudo normal water quality data according to the characteristics of the preprocessed original water quality data, and the discriminator judges the pseudo normal water quality data according to the preprocessed normal water quality data.
Specifically, the steps for training GAN are as follows:
and 2.1, inputting the preprocessed original water quality data containing the abnormity into a generator, and inputting the preprocessed normal water quality data into a discriminator.
Step 2.2, training the generator and the discriminator.
The generator performs learning generation through the original water quality data containing the abnormity, and starts learning generation to obtain pseudo normal water quality data only containing partial water quality data characteristics. The discriminator learns the complete normal water quality data through learning, but still is a partial characteristic at first. And the generator inputs the pseudo-normal water quality data generated by learning into the discriminator to carry out discrimination judgment operation. The pseudo-normal water quality data generated by the generator can not cheat the discriminator, and in the state of mutual game, the pseudo-normal water quality data generated by the generator is closer to the normal water quality data along with continuous training in order to cheat the discriminator. Meanwhile, in order to judge the difference between the pseudo-normal water quality data and the normal water quality data, the discriminator continuously detects the pseudo-normal water quality data through learning, and the discrimination capability is higher and higher. The training is completed when the iterative training is repeated until the 'Nash equilibrium' in the game theory is finally reached, and a generator and a discriminator obtained through the training provide a tool for subsequent abnormal water quality data detection.
And 3, performing anomaly detection on the water quality data to be detected by using the GAN obtained by training in the step 2, comparing the pseudo-normal water quality data generated by the generator with the water quality data to be detected to obtain reconstruction loss, identifying the water quality data to be detected by using the identifier to obtain identification loss, performing accumulation operation on the two losses, and judging whether the water quality data to be detected is anomalous data.
Specifically, the steps of abnormality detection are as follows:
step 3.1: and carrying out PCA operation on the water quality data to be detected to reduce data dimensionality.
Step 3.2: respectively inputting the water quality data to be detected after the data dimension is reduced into a generator and a discriminator, mapping the data in the generator to obtain a hidden space, generating pseudo-normal water quality data by the generator according to random variables in the hidden space, and comparing the pseudo-normal water quality data with the water quality data to be detected to obtain reconstruction loss; and directly carrying out identification judgment operation in the identifier to obtain identification loss.
Step 3.3: and performing simple accumulation operation on the reconstruction loss and the identification loss to obtain a final loss value, judging whether the water quality data is abnormal according to the loss value, and if the loss value is smaller than a preset threshold value, determining the water quality data as normal water quality data, otherwise, determining the water quality data as abnormal water quality data.
The overall implementation of the method of the present invention is illustrated by a specific example.
In one embodiment of the invention, the hardware is a computer configured to include a hardware environment: a CPU: 2 Intel Xeon 6130 processors (2.1GHz/16c)/2666MHz/10.4 GT; GPU: 6 blocks 16G _ TESLA-P100_4096b _ P _ CAC; memory: 16 root 32G ECC Registered DDR 42666; software environment: operating the system: ubantu 16.04; a deep learning framework: tensorflow, pandas, scimit-leern; language and development environment: python 3.6, Anaconda 3, pycharm 2020.
In order to detect abnormal water quality data, the method needs to consider water quality data detected over the years. The original water quality data set used by the invention is from a real-time detection database of a certain water area of inner Mongolia. The acquisition time is 6 months to 11 months in 2019, the acquisition operation points are 9, the data acquisition frequency is 1 hour/time, and the data acquisition indexes are 11. After the water quality detection data is obtained, firstly, preprocessing operation is carried out on the whole data by using a preprocessing method, deletion operation is carried out on the field of the missing data, and then data dimension reduction operation is carried out on the field of the missing data, so that the abnormal detection speed is accelerated, and a pre-training data set is formed. The pre-training data set is then input into the training model. The whole timeGAN training model is divided into two parts, wherein one part is a generator model capable of generating an almost real water quality sequence, and the other part is a discriminator model capable of judging whether the water quality data is normal water quality data. Finally, the two parts are used for carrying out abnormity detection operation, and a reconstruction loss value is obtained by comparing the almost real pseudo water quality data generated by the generator with the water quality data to be detected; and identifying and judging the water quality data to be detected through the identifier to obtain an identification loss value, and finally performing simple accumulation operation on the identification loss value and the water quality data to be detected to obtain a final judgment result. Fig. 2 shows the result of the water quality data abnormality detection in this embodiment. The invention can improve the accuracy of the anomaly detection to a great extent and meet the expected target. Through the example, the water quality abnormal detection method based on the TimeGAN can be seen in consideration of the characteristics of water quality data in the ecological environment, and finally the suspected abnormal water quality data is found by eliminating the interference of irrelevant data.
Although the present invention has been described by way of preferred embodiments, the present invention is not limited to the embodiments described herein, and various changes and modifications may be made without departing from the scope of the present invention.
Claims (4)
1. A water quality abnormal data detection method based on GAN is characterized by comprising the following steps:
step 1, preprocessing original water quality data and normal water quality data, screening and processing irrelevant interference data, and reducing data dimensionality;
step 2, training GAN, wherein the GAN consists of a generator and a discriminator, the generator and the discriminator are in a mutual game state, the generator and the discriminator are trained by utilizing the preprocessed original water quality data and the normal water quality data, the generator generates pseudo normal water quality data according to the characteristics of the preprocessed original water quality data, and the discriminator judges the pseudo normal water quality data according to the preprocessed normal water quality data;
and 3, performing anomaly detection on the water quality data to be detected by using the GAN obtained by training in the step 2, comparing the pseudo-normal water quality data generated by the generator with the water quality data to be detected to obtain reconstruction loss, identifying the water quality data to be detected by using the identifier to obtain identification loss, performing accumulation operation on the two losses, and judging whether the water quality data to be detected is anomalous data.
2. The GAN-based water quality abnormality data detection method as claimed in claim 1, wherein in the step 1, the preprocessing method comprises:
firstly, carrying out primary pretreatment of data replacement and filling on original water quality data or normal water quality data through Python software;
secondly, removing data with missing key fields in the original water quality data or the normal water quality data;
and finally, carrying out PCA operation to reduce the data dimension.
3. The GAN-based water quality abnormality data detection method according to claim 1, wherein the GAN training method in the step 2 is as follows:
sending the preprocessed original water quality data into a generator, sending the preprocessed normal water quality data into a discriminator, learning and generating by the generator to obtain pseudo normal water quality data only containing partial water quality data characteristics, learning by the discriminator to obtain the characteristics of the normal water quality data, inputting the pseudo normal water quality data into the discriminator to perform discrimination and judgment operation, and repeating iteration to finish training to obtain the trained generator and discriminator.
4. The GAN-based water quality abnormality data detection method according to claim 1, wherein in the step 3, the abnormality detection method is as follows:
firstly, carrying out PCA operation on water quality data to be detected to reduce data dimensionality;
secondly, respectively inputting the water quality data to be detected after the data dimension is reduced into a generator and a discriminator, mapping the data in the generator to obtain a hidden space, generating pseudo-normal water quality data by the generator according to random variables in the hidden space, and comparing the pseudo-normal water quality data with the water quality data to be detected to obtain reconstruction loss; in the discriminator, directly carrying out discrimination judgment operation to obtain discrimination loss;
and finally, performing accumulation operation on the reconstruction loss and the identification loss to obtain a final loss value, judging whether the water quality data is abnormal according to the loss value, and if the loss value is smaller than a preset threshold value, determining the water quality data as normal water quality data, otherwise, determining the water quality data as abnormal water quality data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111258369.9A CN114113516A (en) | 2021-10-27 | 2021-10-27 | Water quality abnormal data detection method based on GAN |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111258369.9A CN114113516A (en) | 2021-10-27 | 2021-10-27 | Water quality abnormal data detection method based on GAN |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114113516A true CN114113516A (en) | 2022-03-01 |
Family
ID=80377302
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111258369.9A Pending CN114113516A (en) | 2021-10-27 | 2021-10-27 | Water quality abnormal data detection method based on GAN |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114113516A (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046321A (en) * | 2015-06-25 | 2015-11-11 | 河海大学 | Combined model water level prediction method based on similarity search |
| CN108665106A (en) * | 2018-05-15 | 2018-10-16 | 中国农业大学 | A kind of aquaculture dissolved oxygen prediction method and device |
| CN110598851A (en) * | 2019-08-29 | 2019-12-20 | 北京航空航天大学合肥创新研究院 | Time series data abnormity detection method fusing LSTM and GAN |
| CN111125186A (en) * | 2019-11-25 | 2020-05-08 | 中国科学院计算技术研究所 | Data processing method and system based on questionnaire |
| US20200184106A1 (en) * | 2018-12-10 | 2020-06-11 | Sap Se | Differential Privacy Processing of IoT Streaming Data Using Generative Adversarial Networks |
| CN112215421A (en) * | 2020-10-13 | 2021-01-12 | 北京工业大学 | Deep learning water quality index prediction method based on generation countermeasure network |
| CN112529234A (en) * | 2019-09-18 | 2021-03-19 | 上海交通大学 | Surface water quality prediction method based on deep learning |
| CN112686291A (en) * | 2020-12-24 | 2021-04-20 | 深圳力维智联技术有限公司 | Water quality prediction method, device, system and computer readable storage medium |
| CN112884056A (en) * | 2021-03-04 | 2021-06-01 | 河北工程大学 | Optimized LSTM neural network-based sewage quality prediction method |
| CN112989710A (en) * | 2021-04-22 | 2021-06-18 | 苏州联电能源发展有限公司 | Industrial control sensor numerical value abnormity detection method and device |
| CN113239970A (en) * | 2021-04-16 | 2021-08-10 | 首钢集团有限公司 | Model training method, equipment vibration abnormity detection method and device |
| CN113298297A (en) * | 2021-05-10 | 2021-08-24 | 内蒙古工业大学 | Wind power output power prediction method based on isolated forest and WGAN network |
-
2021
- 2021-10-27 CN CN202111258369.9A patent/CN114113516A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105046321A (en) * | 2015-06-25 | 2015-11-11 | 河海大学 | Combined model water level prediction method based on similarity search |
| CN108665106A (en) * | 2018-05-15 | 2018-10-16 | 中国农业大学 | A kind of aquaculture dissolved oxygen prediction method and device |
| US20200184106A1 (en) * | 2018-12-10 | 2020-06-11 | Sap Se | Differential Privacy Processing of IoT Streaming Data Using Generative Adversarial Networks |
| CN110598851A (en) * | 2019-08-29 | 2019-12-20 | 北京航空航天大学合肥创新研究院 | Time series data abnormity detection method fusing LSTM and GAN |
| CN112529234A (en) * | 2019-09-18 | 2021-03-19 | 上海交通大学 | Surface water quality prediction method based on deep learning |
| CN111125186A (en) * | 2019-11-25 | 2020-05-08 | 中国科学院计算技术研究所 | Data processing method and system based on questionnaire |
| CN112215421A (en) * | 2020-10-13 | 2021-01-12 | 北京工业大学 | Deep learning water quality index prediction method based on generation countermeasure network |
| CN112686291A (en) * | 2020-12-24 | 2021-04-20 | 深圳力维智联技术有限公司 | Water quality prediction method, device, system and computer readable storage medium |
| CN112884056A (en) * | 2021-03-04 | 2021-06-01 | 河北工程大学 | Optimized LSTM neural network-based sewage quality prediction method |
| CN113239970A (en) * | 2021-04-16 | 2021-08-10 | 首钢集团有限公司 | Model training method, equipment vibration abnormity detection method and device |
| CN112989710A (en) * | 2021-04-22 | 2021-06-18 | 苏州联电能源发展有限公司 | Industrial control sensor numerical value abnormity detection method and device |
| CN113298297A (en) * | 2021-05-10 | 2021-08-24 | 内蒙古工业大学 | Wind power output power prediction method based on isolated forest and WGAN network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109492193B (en) | Abnormal network data generation and prediction method based on deep machine learning model | |
| CN107480705B (en) | Tap water pipe water leakage detection method | |
| CN111562108A (en) | Rolling bearing intelligent fault diagnosis method based on CNN and FCMC | |
| CN112529341B (en) | Drilling well leakage probability prediction method based on naive Bayesian algorithm | |
| CN112491891B (en) | Network attack detection method based on hybrid deep learning in Internet of things environment | |
| CN111901340A (en) | Intrusion detection system and method for energy Internet | |
| CN112738014A (en) | Industrial control flow abnormity detection method and system based on convolution time sequence network | |
| CN110334105B (en) | Stream data abnormity detection method based on Storm | |
| CN114168938B (en) | Semi-supervised SQL injection attack detection method based on few abnormal labels | |
| CN115277180B (en) | Block chain log anomaly detection and tracing system | |
| CN114444620A (en) | Indicator diagram fault diagnosis method based on generating type antagonistic neural network | |
| CN112532652A (en) | Attack behavior portrait device and method based on multi-source data | |
| CN108763926B (en) | Industrial control system intrusion detection method with safety immunity capability | |
| CN114113516A (en) | Water quality abnormal data detection method based on GAN | |
| CN116167438A (en) | Transformer fault diagnosis method based on improved quantum genetic algorithm optimized SVM | |
| CN116662899A (en) | Noise-containing data anomaly detection method based on self-adaptive strategy | |
| CN111260029A (en) | Credibility analysis method for air quality data | |
| CN107682302A (en) | Cross-site scripting attack detection method and device | |
| CN113722230B (en) | Integrated evaluation method and device for vulnerability mining capability of fuzzy test tool | |
| CN115913764A (en) | Malicious domain name training data generation method based on generation of countermeasure network | |
| CN114564543A (en) | Carbon footprint acquisition method based on knowledge graph | |
| CN111143436A (en) | Data mining method for big data | |
| CN115378702B (en) | Attack detection system based on Linux system call | |
| CN117527451B (en) | Network intrusion detection method, device, electronic equipment and storage medium | |
| CN116010208A (en) | Log anomaly detection method based on countermeasure migration learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |